Impact of radiation and chemotherapy on risk of dental abnormalities

A report from the Childhood Cancer Survivor Study

Authors

  • Sue C. Kaste DO,

    Corresponding author
    1. Department of Radiological Sciences, St. Jude Children's Research Hospital, Memphis, Tennessee
    2. Department of Oncology, St. Jude Children's Research Hospital, Memphis, Tennessee
    3. Department of Radiology, University of Tennessee School of Health Sciences, Memphis, Tennessee
    • Department of Radiological Sciences, St. Jude Children's Research Hospital, MS 220, 262 Danny Thomas Place, Memphis, TN 38105===

    Search for more papers by this author
    • Fax: (901) 595-3981

  • Pamela Goodman MS,

    1. Cancer Prevention, Fred Hutchinson Cancer Research Center, Seattle, Washington
    Search for more papers by this author
  • Wendy Leisenring ScD,

    1. Cancer Prevention, Fred Hutchinson Cancer Research Center, Seattle, Washington
    2. Clinical Statistics Program, Fred Hutchinson Cancer Research Center, Seattle, Washington
    Search for more papers by this author
  • Marilyn Stovall PhD,

    1. Department of Radiation Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Search for more papers by this author
  • Robert J. Hayashi MD,

    1. Department of Pediatrics, Washington University, St. Louis, Missouri
    Search for more papers by this author
  • Mark Yeazel MD,

    1. Department of Family Medicine and Community Heath, University of Minnesota, Minneapolis, Minnesota
    Search for more papers by this author
  • Soraya Beiraghi DDS,

    1. Department of Developmental and Surgical Science, University of Minnesota, Minneapolis, Minnesota
    Search for more papers by this author
  • Melissa M. Hudson MD,

    1. Department of Oncology, St. Jude Children's Research Hospital, Memphis, Tennessee
    2. Department of Epidemiology and Cancer Control, St. Jude Children's Research Hospital, Memphis, Tennessee
    Search for more papers by this author
  • Charles A. Sklar MD,

    1. Department of Pediatrics, Memorial Sloan Kettering Cancer Center, New York, New York
    Search for more papers by this author
  • Leslie L. Robison PhD,

    1. Department of Epidemiology and Cancer Control, St. Jude Children's Research Hospital, Memphis, Tennessee
    Search for more papers by this author
  • K. Scott Baker MD, MS

    1. Cancer Prevention, Fred Hutchinson Cancer Research Center, Seattle, Washington
    Search for more papers by this author

Abstract

BACKGROUND:

The current study was performed to describe frequencies and risk factors of altered oral health and odontogenesis in childhood cancer survivors.

METHODS:

In total, 9308 survivors who were diagnosed between 1970 and 1986 and 2951 siblings from the Childhood Cancer Survivor Study completed a survey that contained oral-dental health information. The authors analyzed treatment impact, socioeconomic data, and patient demographics on dental outcomes using univariate and multivariate logistic regression models to estimate odds ratios (ORs).

RESULTS:

In multivariate analysis, survivors were more likely to report microdontia (OR, 3.0; 95% confidence interval [95% CI], 2.4-3.8), hypodontia (OR, 1.7; 95% CI, 1.4-2.0), root abnormalities (OR, 3.0; 95% CI, 2.2-4.0), abnormal enamel (OR, 2.4; 95% CI, 2.0-2.9), teeth loss ≥6 (OR, 2.6; 95% CI, 1.9-3.6), severe gingivitis (OR, 1.2; 95% CI, 1.0-1.5), and xerostomia (OR, 9.7; 95% CI, 4.8-19.7). Controlling for chemotherapy and socioeconomic factors, radiation exposure of ≥20 Gray to dentition was associated significantly with an increased risk of ≥1 dental abnormality. Dose-dependent alkylating agent therapy significantly increased the risk of ≥1 anatomic/developmental dental abnormalities in survivors who were diagnosed at age <5 years (OR, 1.7, 2.7, and 3.3 for alkylating agent scores of 1, 2, and 3, respectively).

CONCLUSIONS:

Radiation and chemotherapy were independent risk factors for adverse oral-dental sequelae among childhood cancer survivors. The authors concluded that patients who received receiving alkylating agents at age <5 years should be closely monitored. Cancer 2009. © 2009 American Cancer Society.

Current multimodality therapies have increased the survival of patients with childhood cancer. Many of those therapies are associated with delayed toxicities, resulting in long-term complications, but to our knowledge little has been published to date regarding the effects of pediatric cancer therapy on developing dentition of children. Reported abnormalities include hypodontia (developmentally missing teeth), microdontia (small teeth), enamel hypoplasia, root stunting, taurodontia (enlarged pulp chambers), over-retention of primary teeth, an increased caries index,1-8 malocclusion,9 and decreased temporomandibular joint mobility.10 Local effects of radiotherapy on craniofacial and dental development have been described.1, 2, 11-15 Existing reports have indicated that children may be at greater risks for odontogenic developmental abnormalities if they are treated with chemotherapy at age <5 years because of the proliferation of dental stem cells during this period.14-16 However, the odontogenic toxicities induced by individual chemotherapy agents remain obscure. Thus, the current report includes data available at the time of the analysis and describes the types and frequencies of altered dental development in adult survivors of pediatric cancers and associate disodontogenesis with treatment agents, socioeconomic demographics, and type of malignancy.

MATERIALS AND METHODS

Patient Selection and Contact

The Childhood Cancer Survivor Study is a collaborative, multi-institutional project, funded as a resource by the National Cancer Institute, of individuals who survived ≥5 years after a diagnosis of childhood cancer. For participating investigators, see http://www.stjude.org/ccss (accessed on September 30, 2009).17 Eligibility criteria for this cohort are 1) a diagnosis of leukemia, central nervous system malignancy (all histologies), Hodgkin disease, non-Hodgkin lymphoma, kidney cancer, neuroblastoma, soft tissue sarcoma, or malignant bone tumor (list of eligible International Classification of Diseases for Oncology codes are available at: http://www.stjude.org/ccss; see Supplemental Information to Previous CCSS Publications accessed on September 30, 2009); 2) diagnosis and initial treatment at 1 of the 26 collaborating CCSS institutions; 3) diagnosis date between January 1, 1970 and December 31, 1986; 4) age <21 years at diagnosis; and 5) survival for ≥5 years after diagnosis.

Table 1. Questions Asked of Study Participants Regarding Dental and Oral Abnormalities and Corresponding Abnormality*
  • *

    The Childhood Cancer Survivor Study is a collaborative, multi-institutional project, funded as a resource by the National Cancer Institute, of individuals who survived ≥5 years after a diagnosis of childhood cancer. For participating investigators, see http://www.stjude.org/ccss (accessed on September 30, 2009).

Have you ever…
 Had 1 or more missing teeth because they did not develop?Hypodontia
 Had a lack of or decreased amount of enamel on surface of teeth?Enamel hypoplasia
 Had abnormal shaped (small or malformed) teeth?Microdontia
 Had difficulty producing saliva (dry mouth) that required treatment such as artificial saliva?Xerostomia
 Had severe gingivitis or gum disease requiring surgery or deep cleaning?Gingivitis
 Had more than 5 cavities?> 5 caries
 Lost 6 or more teeth due to decay or gum disease?≥6 Teeth lost
 Worn a dental bridge (for missing or removed teeth)?Dental Bridge
 Worn removable dentures (complete or partial upper or lower or both)?Dentures
 Worn a prosthesis to lift your palate to improve the quality of your voice?Oral prosthesis

The CCSS protocol and contact documents were reviewed and approved by the Human Subjects Committee at each participating institution. Baseline data were collected for the study cohort using a 24-page, self-reported questionnaire. The questionnaire was designed to capture a wide range of information, including demographic characteristics, health habits (smoking, alcohol consumption, physical activity), frequency of diagnosed medical conditions, recurrent cancer, and subsequent primary neoplasms. The medical records of all members of the cohort were abstracted. Detailed data regarding the chemotherapeutic agents administered to the patient for treatment of the original cancer and for any recurrences of the cancer; the cumulative dose of drug administered for several drugs of interest; and the doses, volumes, and dates of administration of all radiotherapy were recorded. Of the 20,626 5-year survivors, 3058 (14.8%) were lost to follow-up and were never offered enrollment. Of the remaining 17,568 survivors, 14,363 (81.8%) completed the baseline survey. Complete medical record abstraction was successful for 12,492 participants. A random sample of participating survivors was selected and asked to contact their sibling closest in age to inform them about the study and invite them to participate. Of the 4790 siblings selected, 3899 (81.4%) completed a baseline survey.

A follow-up survey was distributed beginning in 2003 that updated information on medical conditions and included questions regarding dental health (Table 1; questionnaires are available for review and download at www. stjude.org/ccss accessed on September 30, 2009). Of the 11,723 survivors enrolled in CCSS who were not lost to follow-up, 9308 (79%) completed the 2003 follow-up questionnaire, of whom 8522 had complete treatment data available from the medical record abstraction process and were included in the current analysis (the questions that were used for this study are shown in Table 1). Of the 3899 siblings who were eligible for the follow-up survey; 2951 (85%) completed the second follow-up questionnaire; and, of these, 2831 did not report a cancer event. The current report includes that were data available at the time of the analysis from these 8522 survivors and 2831 siblings.

Statistical Analysis

Descriptive statistics of demographic and treatment characteristics were calculated for the 8522 survivors and 2831 siblings. Self-reported dental outcomes, including hypodontia, microdontia, enamel hypoplasia, abnormal root development, ≥6 missing teeth, denture use, dental prosthesis use, xerostomia, gingivitis, and ≥6 cavities, were considered in this analysis. Three additional outcomes were created from a combination of these individual outcomes and were included in the analyses. The presence of 1 or more of hypodontia, microdontia, enamel hypoplasia, abnormal root development, ≥6 missing teeth, denture use, and dental prosthesis use was defined as “at least 1 dental health issue.” Similarly, the presence of at least 1 of xerostomia, gingivitis, and ≥6 cavities was defined as “at least 1 soft tissue issue”; and the presence of dental bridge use, denture use, or oral prosthesis use defined the outcome “at least 1 appliance use.”

Among the patients who had been exposed to an alkylating agent, the alkylating agent dose score was calculated by adding the tertile score for each of the alkylating agents received by a particular patient.18 Radiation dose to the teeth or teeth buds was estimated for each patient by reviewing and abstracting details of the radiotherapy from radiation oncology records submitted to the Radiation Physics Center at The University of Texas M. D. Anderson Cancer Center. The radiation dose used in the analysis was the mean dose to 12 points throughout the teeth or teeth bud region. Dosimetry methods are described fully elsewhere.19

We conducted a cross-sectional comparison of demographic characteristics and dental outcomes between survivors and siblings using logistic regression models that were adjusted for potential intrafamily correlation with robust sandwich variance estimates.20 Sensitivity analyses were performed to assess the impact of missing or unknown responses to dental outcomes questions, and we assumed that all unknown/missing outcomes were either “no” or “yes” responses. Analyses within the survivor cohort were performed to assess the impact of treatment-related and diagnosis-related factors using individuals who had available treatment data and within the age-at-diagnosis strata birth to 5 years, 6 years to 10 years, and >10 years. Analyses were conducted in SAS statistical software (version 9.1.3; SAS Institute Inc, Cary, NC).

RESULTS

Of the 8522 survivors in the current analysis, 49.9% were female, and 86.8% were white non-Hispanic. The median age at cancer diagnosis was 6 years (age range, birth to 20 years), and the median time from diagnosis to interview was 22 years (range, 15-34 years) (Table 2). The distribution of survivors by diagnosis and details of treatment received are shown in Table 3. The majority of survivors received chemotherapy with or without radiotherapy. Compared with survivors, siblings were significantly more likely to be white, female, aged ≥30 years, high school graduates, to have annual household incomes >$20,000, and to have health insurance or dental insurance.

Table 2. Characteristics of Survivor and Sibling Cohorts
CharacteristicNo. (%)P*
SurvivorsSiblings
  • NA indicates not applicable.

  • *

    P values are from generalized estimating equation models that account for interfamily correlations.

Total cohort85222831 
 History of at least 1 dental health issue
  Yes2539 (32.6)530 (19.9) 
  No5249 (67.4)2132 (80.1)<.01
 History of at least 1 soft tissue issue (includes caries)
  Yes4662 (57.4)1511 (55.8) 
  No3458 (42.6)1197 (44.2).12
 History of dental appliance use
  Yes625 (7.4)168 (6) 
  No7821 (92.6)2642 (94).01
Socioeconomic and demographic characteristics
 Race
  White non-Hispanic7367(86.8)2441 (91.7) 
  Other1124 (13.2)220 (8.3)<.01
 Sex
  Male4273 (50.1)1325 (46.8) 
  Female4249 (49.9)1506 (53.2)<.01
 Age at follow-up, y
  17-293746 (44)1082 (39.2) 
  ≥304776 (56)1675 (60.8)<.01
 Education level
  ≤High school1691 (20.1)489 (17.3) 
  >High school6720 (79.9)2333 (82.7)<.01
 Household income
  <$20,000/y2643 (31.9)635 (22.9) 
  ≥$20,000/y5647 (68.1)2134(77.1)<.01
 Health insurance
  Yes7477 (88.5)2550 (90.4) 
  No971 (11.5)271 (9.6)<.01
 Dental insurance
  Yes5582 (67.6)1963 (70.9) 
  No2680 (32.4)806 (29.1)<.01
Preventive care
 Annual cleaning
  Yes5666 (67.4)1946 (69.4) 
  No2739 (32.6)859 (30.6).05
 Clinic visit within 1 y
  Yes6079 (71.7)2027 (72.1) 
  No2395 (28.3)785 (27.9).72
 Difficulty finding a dentist because of previous cancer
  Yes247 (3) 
  No8091 (97)NA
Table 3. Characteristics of Cancer Diagnosis Treatment
CharacteristicNo. of Survivors (%)
  1. CNS indicates central nervous system; Gy, grays.

Diagnosis
 Leukemia2910 (34.2)
 CNS1076 (12.6)
 Hodgkin disease1086 (12.7)
 Non-Hodgkin lymphoma628 (7.4)
 Kidney (Wilms tumor)794 (9.3)
 Neuroblastoma575 (6.8)
 Soft tissue sarcoma750 (8.8)
 Bone cancer702 (8.2)
Year of diagnosis
 1970-19731078 (12.7)
 1974-19771704 (20)
 1978-19812213 (26)
 1982-19863526 (41.4)
Age at diagnosis, y
 Birth-53971 (46.6)
 6-101661 (19.5)
 ≥112889 (33.9)
Treatment with
 Radiotherapy only1067 (12.5)
 Chemotherapy only2165 (25.4)
 Chemotherapy and radiotherapy4412 (51.8)
 Surgery only involving oral cavity13 (0.2)
 Chemotherapy, radiation, and surgery involving oral cavity124 (1.5)
 Radiation and surgery involving oral cavity17 (0.2)
 Other or unknown724 (8.5)
Alkylating agent score
 04214 (52.7)
 11841 (23)
 21195 (15)
 3740 (9.3)
Cyclophosphamide dose, mg
 04828 (59.1)
 1-3999998 (12.2)
 4000-7999969 (11.9)
 ≥80001374 (16.8)
Antimetabolites
 Yes4063 (47.7)
 No4452 (52.3)
Steroids
 Yes4121 (48.4)
 No4394 (51.6)
Vincristine
 Yes6064 (71.2)
 No2451 (28.8)
Treatment with radiation, Gy
 None2900 (35.2)
 >0 to <205198 (63.1)
 ≥20143 (1.7)

Comparison With the Sibling Cohort

Compared with the sibling cohort, survivors reported a higher frequency of all adverse dental outcomes with the exception of dental bridge use and need for oral prosthesis (Table 4). However, certain differences were more striking than others. In multivariate analyses that were adjusted for sex, race, education, annual household income, health insurance status, and age at study, survivors were statistically significantly more likely than siblings to report dental developmental issues, including microdontia (9.2% vs 3.3%), hypodontia (8.2% vs 5.3%), abnormal root development (5.4% vs 1.9%), enamel hypoplasia (11.7% vs 5.3%), and loss of >6 teeth because of decay or gum disease (4.8% vs 1.8%); and survivors were more likely to have had severe gingivitis (6.7% vs 5.7%). In addition, survivors reported a greater need for dentures (3% vs 1.7%) and an increased frequency of xerostomia (2.8% vs 0.3%). Additional multivariate models that were adjusted for dental insurance and composite dental care use in multivariate models provided virtually identical odds ratios (ORs) (data not shown). Sensitivity analyses were conducted to evaluate the potential impact of missing or unknown outcome responses on associations. Replacement of missing/unknown responses with all “no” responses resulted in no qualitative or marked quantitative differences in P values or point estimates of ORs. The less plausible replacement of missing/unknown with “yes” responses did result in reduced ORs for abnormal roots, microdontia, oral prosthesis, and xerostomia, but there were no changes in the significance of P values.

Table 4. Comparison of Dental Outcomes Between Survivors and Siblings*
Dental CharacteristicNo. of Patients (%)Adjusted OR [95% CI]Adjusted P
Survivor CohortSibling Cohort
YesNoUnknownYesNoUnknown
  • OR indicates odds ratio; 95% CI, 95% confidence interval.

  • *

    Models were adjusted for sex, race, education, household income, health insurance, and age at follow-up, which included generalized estimating equation methods for intrafamily correlations. Additional models, which also included adjustment for dental insurance and dental care use (as a composite), produced virtually no change in observed ORs. Sensitivity analyses to impact the potential impact of unknown responses also were conducted.

>1 Dental abnormality2539 (29.8)5249 (61.6)734 (8.6)530 (18.7)2132 (75.3)169 (6)1.9 [1.7-2.1]<.01
>1 Dental soft tissue abnormality4662 (54.7)3458 (40.6)402 (4.7)1511 (53.4)1197 (42.3)123 (4.3)1.2 [1.1-1.3]<.01
Dental appliance use625 (7.3)7821 (91.8)76 (0.9)168 (5.9)2642 (93.3)21 (0.7)1.3 [1.1-1.6].01
Microdontia785 (9.2)7498 (88)239 (2.8)92 (3.3)2698 (95.3)41 (1.1)3.0 [2.4-3.8]<.01
Hypodontia698 (8.2)7613 (89.3)211 (2.5)149 (5.3)2640 (93.3)42 (1.5)1.7 [1.4-2.0]<.01
>5 Caries4429 (52)3696 (43.4)397 (4.7)1445 (51)1263 (44.6)123 (4.3)1.2 [1.1-1.3]<.01
Abnormal roots464 (5.4)7585 (89)473 (5.5)53 (1.9)2708 (95.7)70 (2.5)3.0 [2.2-4.0]<.01
Enamel hypoplasia998 (11.7)6706 (79.7)818 (9.6)151 (5.3)2535 (89.5)145 (5.1)2.4 [2.0-2.9]<.01
≥6 Teeth lost410 (4.8)8030 (94.2)82 (1)51 (1.8)2768 (97.8)12 (0.4)2.6 [1.9-3.6]<.01
Gingivitis572 (6.7)7808 (1.6)142 (1.7)161 (5.7)2647 (93.5)23 (0.8)1.2 [1.0-1.5].03
Xerostomia237 (2.8)8203 (96.3)82 (1)8 (0.3)2813 (99.4)10 (0.4)9.7 [4.8-19.7]<.01
Dental bridge458 (5.4)8003 (93.9)61 (0.7)140 (5)2675 (94.5)16 (0.6)1.2 [1.0-1.4].12
Dentures252 (3)8224 (96.5)46 (0.5)47 (1.7)2774 (97.8)10 (0.4)1.7 [1.2-2.4]<.01
Oral prosthesis30 (0.4)8430 (98.9)62 (0.7)2 (0.1)2815 (99.4)14 (0.5)3.9 [0.9-16.7].07

Use of Dental Services

Acknowledging that the reporting of many dental abnormalities may depend in part on their diagnosis by a dental healthcare professional, we examined associations between the use of dental services and socioeconomic factors. We observed no significant difference in the use of dental services between survivors and siblings (Table 2) but significantly higher frequencies of both cleaning visits and seeing a dentist in the past year for individuals who were white, those with an educational level beyond high school, those with annual income levels ≥$20,000, and female (Table 5).

Table 5. Survivors' Use of Dental Services
VariableNo. of Patients (%)OR95% CIP
YesNo
  1. OR indicates odds ratio; 95% CI, 95% confidence interval.

Teeth cleaned in last year
 Education
  >High school4655 (83.3)1981 (73.2)1.81.6-2.0<.01
  ≤High school936 (16.7)727 (26.9)1.0  
 Sex
  Female2986 (52.7)1213 (44.3)1.41.3-1.5<.01
  Male2680 (47.3)1526 (55.7)1.0  
 Household income
  ≥$20,0004123 (74.7)1456 (54.7)2.52.2-2.7<.01
  <$20,0001396 (25.3)1207 (45.3)1.0  
 Race
  White non-Hispanic4942 (87.5)2323 (85.1)1.21.1-1.4<.01
  Other704 (12.5)406 (14.9)1.0  
Visit to dentist in last year
 Education
  >High school4938 (82.4)1755 (74)1.61.5-1.8<.01
  ≤High school1057 (17.6)617 (26)1.0  
 Sex
  Female3193 (52.5)1038 (43.3)1.51.3-1.6<.01
  Male2886 (47.5)1357 (56.7)1.0  
 Household income
  ≥$20,0004320 (73.1)1301 (55.7)2.22.0-2.4<.01
  <$20,0001591 (26.9)1037 (44.4)1.0  
 Race
  White non-Hispanic5291 (87.3)2032 (85.2)1.21.1-1.4<.01
  Other766 (12.7)354 (14.8)1.0  

Risk Factors for Dental Abnormalities

To examine the influence of demographic factors (sex, race/ethnicity), socioeconomic factors (education, income, insurance status), and the impact of treatment exposure (radiation and chemotherapy, including alkylating agents, antimetabolites, steroids, and vincristine), we performed a multivariate logistic regression analysis (Tables 6, 7). A low frequency of events prohibited analysis of some outcomes, including the need for oral appliance use. Increased risk of at least 1 reported dental health issue was associated with younger age at study, female, white non-Hispanic race, educational attainment of high school or less, and an annual household income <$20,000. With the exception of age at study and educational attainment level, similar risks were observed for at least 1 reported soft tissue issue. The dose of radiation to the teeth had a statistically significant association with an increased risk of 1 or more dental health issues (exposure to doses >0 grays [Gy] and <20 Gy: OR, 1.3; exposure to doses ≥20 Gy: OR, 5.6) and 1 or more soft tissue issues (exposure to doses >0 Gy and <20 Gy: OR, 1.2; exposure to doses ≥20 Gy: OR, 2.2). Similarly, increasing cumulative exposure to alkylating agent therapy was associated with an increased risk of reporting at least 1 dental abnormality (Table 7).

Table 6. Odds Ratios for Socioeconomic Variables: Multivariate Logistic Regression Model Adjusted for Radiation Dose, Alkylating Agents Score, Steroids, Vincristine, and Antimetabolites
VariableAt least 1 Dental Health IssueAt Least 1 Soft Tissue IssueDental ApplianceMicrodontiaHypodontia≥6 CavitiesAbnormal RootsEnamel HypoplasiaLost ≥6 TeethXerostomia
OR95% CIOR95% CIOR95% CIOR95% CIOR95% CIOR95% CIOR95% CIOR95% CIOR95% CIOR95% CI
  • OR indicates odds ratio; 95% CI, 95% confidence interval.

  • *

    .01 < P < .05.

  • P ≤ .01.

Age at follow-up, y
 <301.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 
 ≥300.9*0.8-1.02.62.4-2.92.62.1-3.20.40.3-0.50.60.5-0.72.42.2-2.70.50.4-0.60.90.8-1.12.72.1-3.52.71.8-3.9
Sex
 Male1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 
 Female1.31.2-1.41.41.3-1.61.21.0-1.41.2*1.0-1.41.61.3-1.91.51.3-1.61.10.9-1.41.31.1-1.51.10.9-1.41.10.8-1.5
Race/ethnicity
 Other1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 
 White, non-Hispanic1.21.0-1.41.21.0-1.41.4*1.0-1.91.10.9-1.41.41.1-1.91.21.1-1.41.41.0-1.91.41.1-1.71.10.8-1.51.50.9-2.4
Education
 ≤High school1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 
 >High school0.80.7-0.91.10.9-1.20.8*0.6-1.00.70.5-0.80.90.8-1.21.11.0-1.20.70.5-0.90.8*0.7-1.00.40.3-0.51.6*1.0-2.5
Household income
 <$20,0001.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 
 ≥$20,0000.80.7-0.90.9*0.8-1.00.80.7-1.00.80.7-0.91.10.9-1.30.9*0.8-1.00.70.6-0.90.70.6-0.80.50.4-0.60.60.5-0.8
Health insurance
 No insurance1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 
 Yes or Canadian0.90.7-1.00.90.8-1.10.90.7-1.21.10.9-1.41.10.8-1.40.90.8-1.10.90.6-1.20.8*0.6-1.00.70.5-0.91.00.6-1.6
Table 7. Odds Ratios for Radiation and Chemotherapy Exposure in a Multivariate Logistic Regression Model Adjusting for Age at Follow-Up, Sex, Race/Ethnicity, Education, Household Income, and Health Insurance
VariableAt Least 1 Dental Health IssueAt Least 1 Soft Tissue Issue
OR95% CIPOR95% CIP
  1. OR indicates odds ratio; 95% CI, 95% confidence interval; Gy, Gray.

Radiation dose to the jaw, Gy
 01.0  1.0  
 >0 to <201.31.2-1.5<.011.21.1-1.3<.01
 ≥205.63.7-8.5<.012.21.4-3.4<.01
Cumulative alkylating agent dose score
 01.0  1.0  
 11.41.2-1.6<.011.00.9-1.1.77
 21.71.5-2.0<.011.41.2-1.6<.01
 32.01.6-2.4<.011.31.1-1.6<.01
Antimetabolite therapy
 No1.0  1.0  
 Yes1.00.9-1.2.890.90.8-1.1.43
Steroid therapy
 No1.0  1.0  
 Yes0.90.8-1.1.181.11.0-1.3.13
Vincristine therapy
 No1.0  1.0  
 Yes1.00.8-1.1.680.90.8-1.0.19

Treatment factors were examined in these same multivariate models for all ages combined and stratified by age at cancer treatment. Radiotherapy to the teeth significantly increased the risk of developing 1 or more dental abnormalities in a dose-dependent pattern (doses >0 Gy and <20 Gy: OR, 1.3 [95% confidence interval (95% CI), 1.2-1.5]; doses ≥20 Gy: OR, 5.6 [95% CI, 3.7-8.5]). This dose-dependent risk was present within all age strata, except for survivors who were aged >10 years at diagnosis who received doses >0 Gy and <20 Gy (ages birth to 5 years: OR, 1.3 [95% CI, 1.2-1.5] for doses >0 Gy and <20 Gy and OR, 5.6 [95% CI, 3.7-8.5] for doses ≥20 Gy; ages 6-10 years: OR, 1.5 [95% CI, 1.1-1.9] for doses >0 Gy and <20 Gy and OR, 9.6 [95% CI, 4.1-22.4] for doses ≥20 Gy; and age >10 years: OR, 1.2 [95% CI, 0.9-1.4] for doses >0 Gy and <20 Gy and OR, 4.3 [95% CI, 2.2-8.3] for doses ≥20 Gy). Investigation of the risk for specific dental issues according to cumulative exposure to alkylating agent therapy and for cumulative dose of cyclophosphamide demonstrated a striking association with increased risk, independent of radiation exposure to the teeth, among survivors who were diagnosed and treated for their cancer when they were aged <5 years (Fig. 1).

Figure 1.

Risk of dental abnormalities by alkylating agent exposure is shown. Solid circles represent alkylating agent score as described in the methods with referent as a score of 0. Open circles represent cumulative cyclophosphamide exposure in grams, with referent as no cyclophosphamide. Odds ratios and 95% confidence limits are adjusted for age at follow-up, sex, race, education, household income, insurance, radiation exposure to the jaw, steroids, antimetabolites, and vincristine.

Although radiation exposure to the teeth and alkylating agent therapy were found to be independent risk factors for dental abnormalities, analyses were conducted to determine whether there was a potential interaction effect. These tests indicated that only the outcomes of microdontia (P = .01) and enamel hypoplasia (P = .02) had a statistically significant, interactive effect between radiation to the teeth and the use of alkylating agents.

DISCUSSION

Delayed toxicities resulting from modern cancer therapy are recognized increasingly, and many have the potential to compromise long-term health. Through the CCSS, with the large number of participants and detailed treatment information, we identified dose-associated and age-associated exposure to alkylating agents, independent of radiation exposure, as 1 of the primary causes of adverse dental outcomes in childhood cancer survivors.

Animal studies have demonstrated that alkylating agents, and specifically cyclophosphamide, are toxic to normal dentinogenesis by their binding to DNA in the S-phase of mitosis, ultimately resulting in early apoptosis.21-23 The dental effect of cyclophosphamide-induced cell death predominates in primitive mesenchymal cells and preodontoblasts of the pulp.24 Resulting outcomes may reflect the stage of dental development that was occurring at the time of exposure to this agent, such that, the younger the animal, the greater the effect on dentition. Upon completion of dental development, mature teeth, particularly incisors and molars, may exhibit a variety of abnormalities, including foreshortened root development; small, soft crowns, abnormal, partially calcified pulp chambers; misshaped teeth; and enamel defects.25-28 Although findings from animal studies have been supported through case reports and investigations among small cohorts of pediatric cancer patients,25, 29 the current CCSS study definitively identifies young age and increased exposure to alkylating agents as risk factors for developmental dental abnormalities in long-term survivors of childhood cancer.

Altered odontogenesis and exfoliation may be accompanied by altered dental eruption. In rat studies, corticosteroids accelerated and cyclophosphamide slowed the eruption of rat incisors during the normal phase of eruption; whereas neither drug affected eruption during the initial, slow phase.30 Because the rapid growth of rat dentition has been likened to the rapid dental growth and development in children,23 similar effects also may be expected to manifest in children who are exposed to this agent. In fact, a patient is at greater risk for odontogenic developmental abnormalities if they are treated with chemotherapy at ages <5 years because of the prolific activity of dental stem cells during this period.14-16 Our data emphasize the significant vulnerability of developing dentition in young children when they receive alkylator therapy. When either cyclophosphamide or the alkylator index was examined, a dose response was demonstrated that indicated a greater frequency of adverse dental outcomes with higher exposure to alkylator chemotherapy. Children aged >5 years failed to demonstrate these findings even at the highest levels of alkylator exposure.

Radiotherapy may alter dental integrity1, 2 and craniofacial development.12, 31 Xerostomia is a common side effect during radiation treatment32 and contributes to altered oral flora, which, in turn, can be associated with an increased number of caries.33, 34 It also has been demonstrated that radiotherapy damages the tooth bud, thereby causing microdontia, growth retardation of teeth, malocclusion, and arrested root development.1, 2, 35, 36 Atrophy of underlying soft tissue, enamel hypoplasia, or incomplete calcification also can result. The degree and severity of these effects depend on the child's age at diagnosis and the type and dose of radiation,37-39 as supported by our current findings. Sonis et al observed that acute lymphoblastic leukemia survivors who received 24 Gy of radiation were affected more severely than those who received 18 Gy.16 We also observed that exposure to radiotherapy doses >20 Gy contributed to a 4-fold to 10-fold higher risk of developing dental abnormalities.

Furthermore, we observed that sex and socioeconomic factors were correlated with odontogenic toxicity after adjusting for treatment differences and also were associated with significant differences in the use of healthcare.40 Female sex, white non-Hispanic race, lower education level, and lower household income level were associated with an increased risk of adverse dental health in childhood cancer survivors. In contrast, patients with higher incomes and more education actually reported fewer dental abnormalities. An increased risk of dental abnormalities may reflect decreased access to dental care (particularly limited access to dentists trained in caring for these complex patients) and, potentially, decreased use of preventive care. Individuals of white race, those with educational level beyond high school, income levels ≥$20,000, and females all reported a greater frequency of having cleaning visits and seeing a dentist in the last year. Such an increase in dental care use may result in a greater likelihood of diagnosis and, hence, knowledge of specific dental conditions rather than reflecting a difference in biologic susceptibility. These patients who presumably have better access to dental care and possibly greater health awareness also may have received preventative dental care that minimized some adverse outcomes.

In interpreting the results of the current study, it is important to consider its limitations and strengths. The validity of self-reported dental abnormalities in this population is unknown. The ability to self-report many abnormalities described in this report may be dependent on the level of dental care. Missing teeth, the number of cavities, and the use of dental appliances are readily apparent to the individual. However, the recognition of abnormalities, such as lack of enamel (enamel hypoplasia), small teeth (microdontia), and root abnormalities (root stunting), requires a professional diagnosis. Strengths of this study include the large survivor and sibling cohorts, known cumulative radiation and chemotherapy dose exposures, and long-term follow-up, which allowed us to conduct a detailed analysis of the factors that contribute to dental developmental abnormalities.

Our data indicate that children who are aged <5 years when they are exposed to alkylating agents, particularly those who receive high cumulative doses, are at high risk for developmental dental abnormalities. Although this may not be avoidable for many patients given the wide use of alkylating agents in pediatric oncology, these data can be used to inform parents, patients, and dental care providers that close monitoring and follow-up of these children will be necessary. The same caveats hold true for children who are exposed to radiotherapy to the teeth, particularly if the dose is >20 Gy. The need for ongoing, close dental follow-up should be conveyed to cancer survivors and their parents during therapy and should be re-emphasized upon completion of therapy.

Acknowledgements

We thank Sandra Gaither for article preparation.

Conflict of Interest Disclosures

Supported by grant U24-CA-55727 (L. L. Robison, principal investigator) from the Department of Health and Human Services, funding to the University of Minnesota from the Children's Cancer Research Fund, and funding to St. Jude Children's Research Hospital from the American Lebanese Syrian Associated Charities.

Ancillary